active-directory-attacks
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute various offensive tools such as Impacket, BloodHound, Mimikatz, Rubeus, and CrackMapExec. These commands involve network operations and local process execution aimed at Active Directory targets.
- [PRIVILEGE_ESCALATION]: The documentation describes several methods for gaining elevated domain privileges, including DCSync attacks, Golden/Silver ticket forging, and exploitation of misconfigured Certificate Services (AD CS) templates. These techniques are explicitly designed to obtain Domain Administrator access.
- [INDIRECT_PROMPT_INJECTION]: The skill maintains an attack surface for indirect prompt injection through its data ingestion workflows.
- Ingestion points: The agent is instructed to parse and act upon output from Active Directory enumeration tools (e.g., BloodHound, PowerView) and network scanning utilities.
- Boundary markers: A mandatory confirmation gate is included in the preamble, requiring user verification of targets and authorization before any action is taken.
- Capability inventory: The skill utilizes shell execution (bash/powershell), network communication via specialized tools, and credential harvesting capabilities.
- Sanitization: No explicit sanitization or validation of data ingested from domain environments or tool outputs is provided.
Audit Metadata